由於無線感測網路（WSNs）環境的開放性，造成容易受到惡意節點的內部攻擊，現有的大多數研究都計算感測器之間的成功率來代表節點的直接信任，使用信任評估模型是檢測未知節點和保證無線感測網路安全的有效方法。然而，如同開關攻擊這種內部惡意行為，在只計算兩個節點間的直接信任模型下，是不足以確保彼此節點是可信的。因此本文提出了一種新穎的信任模型，該模型考慮節點通訊和剩餘電量來計算感測器節點之間的直接信任，並收集鄰居節點的意見來計算間接信任。此外，我們也解釋了直接信任和間接信任之間的權重來做信任值的整合，最後，此模型包含一個基於滑動窗口方法的加權平均來定期更新最終信任值。另外，我們開發了一種基於可靠信任的拓樸控制演算法 (RTTC)，以利用我們的信任模型並創建基於安全性的網路拓樸設計。基於模擬結果，我們提出的模型可以準確評估感測器節點的可信度，有效防止安全漏洞。;Wireless sensor networks (WSNs) are vulnerable to internal attacks because of the openness of network environments. Using a trust evaluation model is a helpful approach for detecting unidentified sensor nodes and ensuring security protection in WSNs. However, most existing studies consider the communications between nodes to calculate the direct trust, which may not be enough to compute the trust value due to various internal attacks like on-off attacks and malicious behavior. This thesis proposes a novel trust model that considers communication and residual energy to calculate direct trust between sensor nodes and uses neighbors′ direct trusts to evaluate indirect trusts. Then, the weight between direct and recommendation trusts is explained. Finally, the integrated trust′s value is updated periodically so that this model includes a weighted average operator based on the sliding window method. Furthermore, we developed a reliable trust-based topology control algorithm (RTTC) to utilize our trust model and create a security-based network topology. Based on simulation results, our proposed model can accurately assess sensor nodes′ trustworthiness and effectively prevent security breaches.
